This project has sought to exploit genetic manipulation of the mouse genome in order to study the development and characteristics of the cells comprising the pancreatic islets. Our proposition has been that such information will contribute to basic knowledge of the pancreas and its diseases (not only diabetes but also cancer), and perhaps to the eventual design of strategies to replenish the beta cells through the control of beta cell growth and differentiation. A major new focus of this project is applying to diabetes research the technology of homologous integration into embryonic stems cells so as to produce mice with targeted gene disruptions (or knockouts). A second new focus is on an exciting cell type associated with pancreatic islets that may represent a multipotential 'stem-like' cell for islet neogenesis and pancreatic cancer. Our specific aims are to: 1. Produce mice carrying a targeted knockout of the gene encoding the autoantigen glutamic acid decarboxylase (GAD65), discovered by Project 2 during the last funding period, and assess by its absence the functional contributions of GAD65 to normal development and postnatal activities of the pancreatic islets. 2. Produce mice carrying a knockout of betacellulin, a TGF/alpha family member implicated in beta cell tumorigenesis, so as to investigate its involvement in islet cell development and islet morphogenesis, as well as physiological function in the adult. 3. Homologously integrate the oncogene SV40 T-antigen into the somatostatin and pancreatic polypeptide genes in embryonic stem cells, and derive mice that carry the modified SMS/Tag and PP/Tag alleles with consequent d and PP cell tumorigenesis, from which representative tumor cell lines will be established. The knockout mice will also be bred to homozygosity, to assess the effects that absence of somatostatin and pancreatic polypeptide have on embryogenesis, as well as islet development, morphogenesis, and function. 4. Introduction of polyomavirus mT oncogene into juvenile islets resulted in establishment of epitheloid cell lines (mPAC) that heterogeneously co-express a pancreatic duct cell marker, cytokeratin, and a neuroendocrine marker, A2B5, in addition to somatostatin and pancreatic polypeptide mRNAs. Upon transplantation, mPAC cells produce classic ductal adenocarcinomas analogous to the major human pancreatic cancer. The cell type from which mPAC originate within juvenile islets, the cellular characteristics of mPAC, and the signaling pathway used in their transformation will be studied to clarify the possibility that mPAC cells represent on the one hand a duct cell progenitor of the islets, and on the other a target cell for pancreatic cancer.